Offgridders not compatible with newest numpy version

[smartie2076]

Installing Offgridders in a new environment and running it leads to following bug:

(D:\PycharmProjects\offgridders_ws\venv) D:\PycharmProjects\offgridders_ws>python Offgridders.py inputs/test_input_template.xlsx
11:29:41-INFO-Path for logging: ./micro_grid_design_logfile.log
11:29:41-INFO-
 Coded by: Martha M. Hoffmann
 Reiner Lemoine Institute (Berlin)


11:29:41-INFO-Performing simulations defined by file "inputs/test_input_template.xlsx"

11:29:41-INFO-Following project locations are evaluated: test_site
11:29:42-WARNING-Attributes "pv_cost_investment, pv_lifetime" defined in constant and sensitivity parameters. Only sensitivity parameter value will be used for sensitivity_exper
iment_s.
11:29:42-INFO-Parameters of sensitivity analysis: pv_cost_investment, pv_lifetime
11:29:42-INFO-NumExpr defaulting to 8 threads.
11:29:42-ERROR-You chose parameter 'fuel_price_change_annual' unequal zero. This calculation is still faulty and you should check the resulting fuel price. It would be better if
 you set your fuel price by hand.
11:29:42-INFO- The resulting fuel price is: 1.044592173358807
11:29:42-INFO-Randomized blackout timeseries for all combinations of blackout duration and frequency (1 experiments) will be generated.
Traceback (most recent call last):
  File "Offgridders.py", line 4, in <module>
    main()
  File "D:\PycharmProjects\offgridders_ws\src\cli.py", line 109, in main
    settings, parameters_constant_values, parameters_sensitivity, project_site_s
  File "D:\PycharmProjects\offgridders_ws\src\C_sensitivity_experiments.py", line 373, in get
    settings, number_of_project_sites, sensitivity_array_dict
  File "D:\PycharmProjects\offgridders_ws\src\C_sensitivity_experiments.py", line 1224, in overall_results_title
    title_overall_results = pd.DataFrame(columns=[CASE, PROJECT_SITE_NAME])
  File "D:\PycharmProjects\offgridders_ws\venv\lib\site-packages\pandas\core\frame.py", line 348, in __init__
    mgr = self._init_dict(data, index, columns, dtype=dtype)
  File "D:\PycharmProjects\offgridders_ws\venv\lib\site-packages\pandas\core\frame.py", line 451, in _init_dict
    nan_dtype)
  File "D:\PycharmProjects\offgridders_ws\venv\lib\site-packages\pandas\core\dtypes\cast.py", line 1196, in construct_1d_arraylike_from_scalar
    dtype = dtype.dtype
AttributeError: type object 'object' has no attribute 'dtype'

Uninstalling numpy==1.21.2 and installing numpy==1.19.4 fixed the issue.